The marketplace cold start problem: supply or demand first?
A marketplace has a chicken-and-egg problem: sellers will not come without buyers, buyers without listings. Here are proven ways to solve the cold start, with the honest condition under which none of them work.
You solve a marketplace cold start by seeding one side of the market manually and giving it value before the network exists. In practice it is usually easier to start with supply (a controlled group of sellers you win through relationships) and attract demand with an offering that is valuable even with few sellers. There is no single right answer to "supply or demand first", because it depends on your advantage. Below are the strategies that work, and one condition under which none of them will.
The heart of the problem is a closed loop: each side of the market waits for the other, so without intervention nothing moves.
Why this is a chicken-and-egg problem
A marketplace only has value once both sides are present: sellers come for buyers, buyers for listings. At the start there is neither, so each side has a rational reason not to come first. This is not a marketing problem to solve with an ad budget but a structural one: as long as the loop is empty, paid traffic bounces off a platform with no listings and returns to the competition. That is why a marketplace is three businesses at once, and the hardest of them is starting that loop.
Strategy 1: seed one side by hand
The most effective and least glamorous method: do not scale, win your first sellers one by one, through relationships and legwork. Twenty well-chosen sellers with a real catalog give buyers a reason to come, and you keep control over the quality of the first supply. This does not scale, and that is the point: hand-seeding one side is an investment, not the end-state process. Many well-known marketplaces started with manual, unscalable work on one side of the market, long before any automation. Automation comes after the loop starts, not before.
Strategy 2: be useful to one side without the other
The second method, stronger if it fits your model: give one side value that exists even with zero of the other side. A tool that helps a seller manage listings or settlements is useful even when there are no buyers yet. A catalog or data a buyer needs regardless of the number of sellers attracts them before a transaction exists. This strategy turns a cold start into a warmer one: one side has a reason to stay even though the network is not there yet, and the network grows on that base.
Strategy 3: niche by niche, not the whole market at once
The third method concerns scope. Liquidity is easier to reach in a small, dense market than a large, dispersed one. Instead of opening a marketplace for every category across a whole country, launch one category in one city or one narrow product niche. The density of supply and demand in a narrow slice makes the loop close faster, and success in one niche gives you proof and material to enter the next. Spreading across the whole market at once is the most common way marketplaces dilute liquidity and die of thinness.
The condition under which no strategy helps
There is one situation in which a cold start cannot be solved by any of the above: the absence of an unfair advantage in reaching either side of the market. If you have no brand, community, industry relationships or a tool that delivers value on its own, then software alone will not start liquidity, no matter how good it is. This is the most important honest note in this article: marketplace technology is a solvable problem, and liquidity is not a technology problem. Before you build the platform, answer why the first side of the market would come to you specifically.
Takeaway: a cold start is not an ad-budget problem but a question of which side of the market you can seed first and why you specifically. Without an unfair advantage, no strategy starts liquidity.
How to turn this into a launch
A practical order: first name your advantage (what makes you reach supply or demand more cheaply than others), then choose the side to seed by hand and the narrowest niche where density is achievable. Only with that answer should you build the platform, and in an MVP version that proves liquidity before you add features. We describe what that smallest sensible marketplace looks like on our marketplace MVP service page.
Supply or demand: three questions that settle it
You cannot say in advance which side to start with, but three questions usually settle it. First: which side do you reach more cheaply because of who you are? A brand and community pull demand; industry relationships and a supplier base pull supply. Second: which side brings value by being first? It is usually easier to give a seller value with a tool than a buyer with an empty catalog. Third: which side is more patient? Sellers tolerate few buyers longer than buyers tolerate an empty offering. The answers rarely point to demand, which is why most successful marketplaces start by seeding supply in a controlled way.
A cold start is not a problem to solve with an ad budget or better design. It is a question of which side of the market you can seed first and why you specifically. Marketplaces that have that question well answered start the loop. The rest build a beautiful, empty platform.
FAQ
Should you start with supply or demand?
Usually supply is easier, a controlled group of sellers won through relationships. Ultimately it depends on which side you reach more cheaply because of who you are.
How many sellers are enough to start?
It is not about the number but density in a narrow niche. A dozen or so well-chosen sellers with a real catalog give buyers a reason to come.
Can liquidity be bought with ads?
No. As long as the loop is empty, paid traffic bounces off a platform with no listings and returns to the competition. Liquidity is not a technology problem.
When should you build the marketplace platform?
Only after naming your advantage and choosing the narrowest niche, in an MVP that proves liquidity before you add features. We describe it on our marketplace MVP service page.
Journal
Co-founder of Seedlight · founder of Amazonway, marketplace sales
Newsletter
The Journal, straight to your inbox
New articles and lessons from real builds, every now and then. No spam, unsubscribe with one click.